Skip to main content

Table 5 Mediation effect of HOMA-IR on the relationship between glucose metabolism–relevant SNPs and CRC risk, stratified by obesity status and obesity-related factors

From: Effect of genetic variants and traits related to glucose metabolism and their interaction with obesity on breast and colorectal cancer risk among postmenopausal women

SNP Nearest gene Effect allele/ Other allele Favorable Energy Balance Group Unfavorable Energy Balance Group
Direct effect Indirect effect Total effect Direct effect Indirect effect Total effect
CRC risk in relation to SNP through pathways other than HOMA-IR CRC risk in relation to SNP through HOMA-IR CRC risk in relation to SNP CRC risk in relation to SNP through pathways other than HOMA-IR CRC risk in relation to SNP through HOMA-IR CRC risk in relation to SNP
HRa 95% CI Effect sizea 95% CI HRa 95% CI HRa 95% CI Effect sizea 95% CI HRa 95% CI
BMIb
 rs2191349 DGKB/TMEM195 G/T 1.04 (0.84–1.28) <0.001 (−0.0040.004) 1.07 (0.90–1.28) 1.45 (1.03–2.03) <0.001 (−0.0030.003) 1.17 (0.90–1.52)
 rs4607517 GCK G/A 0.79 (0.61–1.03) <0.001 (−0.01–0.01) 0.80 (0.64–1.00) 1.04 (0.66–1.62) <0.001 (−0.0040.003) 1.06 (0.75–1.51)
 rs174550 FADS1 T/C 1.16 (0.93–1.45) <0.001 (−0.0010.001) 1.05 (0.87–1.27) 1.40 (0.96–2.04) −0.01 (−0.020.01) 1.37 (1.02–1.83)
 rs11605924 CRY2 C/A 0.87 (0.71–1.07) <0.001 (−0.01–0.01) 0.82 (0.69–0.98) 1.32 (0.93–1.87) <0.001 (−0.0030.003) 1.07 (0.82–1.39)
Waistc
 rs2191349 DGKB/TMEM195 G/T 0.97 (0.76–1.23) 0.002 (−0.01–0.01) 1.00 (0.81–1.23) 1.37 (1.05–1.79) −0.002 (−0.004–0.01) 1.20 (0.97–1.48)
 rs174550 FADS1 T/C 1.22 (0.94–1.58) −0.002 (−0.01–0.01) 1.03 (0.83–1.28) 1.24 (0.93–1.65) −0.01 (−0.02–0.002) 1.25 (1.00–1.57)
 rs11605924 CRY2 C/A 0.84 (0.67–1.06) <0.001 (−0.01–0.01) 0.81 (0.66–0.99) 1.17 (0.90–1.54) <0.001 (−0.01–0.01) 0.97 (0.79–1.20)
Waist/hip Ratiod
 rs2191349 DGKB/TMEM195 G/T 1.04 (0.84–1.29) 0.003 (−0.01–0.004) 1.03 (0.86–1.23) 1.38 (1.00–1.89) <0.001 (−0.01–0.01) 1.24 (0.97–1.59)
 rs10885122 ADRA2A G/T 0.78 (0.58–1.05) <0.001 (−0.01–0.01) 0.84 (0.64–1.08) 1.31 (0.77–2.20) <0.001 (−0.01–0.01) 1.58 (1.01–2.48)
 rs174550 FADS1 T/C 1.15 (0.92–1.45) 0.002 (−0.004–0.01) 1.08 (0.89–1.30) 1.43 (1.01–2.03) −0.008 (−0.02–0.004) 1.27 (0.97–1.66)
Physical activity levele
 rs4607517 GCK G/A 0.73 (0.52–1.01) <0.001 (−0.003–0.003) 0.72 (0.55–0.95) 0.95 (0.69–1.31) −0.001 (−0.01–0.004) 0.99 (0.76–1.28)
Dietary fat intakef
 rs4607517 GCK G/A 0.92 (0.72–1.18) <0.001 (−0.01–0.004) 0.94 (0.76–1.17) 0.58 (0.33–1.03) −0.001 (−0.01–0.001) 0.66 (0.43–1.00)
 rs11558471 SLC30A8 A/G 1.08 (0.88–1.33) <0.001 (−0.01–0.01) 0.95 (0.80–1.12) 1.64 (0.94–2.85) −0.003 (−0.01–0.01) 1.60 (1.07–2.40)
  1. BMI body mass index, CI confidence interval, CRC colorectal cancer, HOMA-IR homeostatic model assessment–insulin resistance, HR hazard ratio, SNP single–nucleotide polymorphism, w/h ratio waist-to-hip ratio
  2. Note: Proportions explained by HOMA-IR for SNP–CRC risk association for rs2191349 (42.8%, 2.0%, and 33.3% among non-obese group [BMI < 30, waist ≤88 cm, w/h ≤ 0.85, respectively]; N/A [> 100%], 85.0%, and 58.3% among obese-group [BMI ≥ 30, waist >88 cm, w/h > 0.85, respectively]), for rs4607517 (0.4%, 0.7%, and 2.7% among non-obese group [BMI < 30, MET ≥10, and <40% calories from fat, respectively]; 42.8%, 3.6%, and 10.9% among obese-group [BMI ≥ 30, MET <10, and ≥40% calories from fat, respectively]), for rs174550 (N/A [> 100%], N/A [> 100%], and N/A[> 100%] among non-obese group [BMI < 30, waist ≤88 cm, and w/h ≤ 0.85, respectively]; 8.7%, 6.3%, and 63.3% among obese-group [BMI ≥ 30, waist >88 cm, and w/h > 0.85, respectively]), for rs11605924 (6.1% and 3.7% among non-obese group [BMI < 30 and waist ≤88 cm, respectively]; N/A [> 100%] and N/A [> 100%] among obese-group [BMI ≥ 30 and waist >88 cm, respectively]), for rs10885122 (6.7% in w/h ≤ 0.85; 47.6% in w/h > 0.85), and for rs11558471 (13.8% in <40% calories from fat; 5.7% in ≥40% calories from fat). Only SNPs having statistically significant results are included. Numbers in bold face are statistically significant
  3. aMultivariate regression was adjusted by covariates (age, education, family history of diabetes mellitus, family history of colorectal cancer, cardiovascular disease ever, hypertension ever, high cholesterol requiring pills ever, total Healthy Eating Index-2005 score, dietary alcohol and total sugars per day, smoking status, lifetime partner, depressive symptom, oral contraceptive use, history of hysterectomy or oophorectomy, age at menarche, age at menopause, pregnancy history, breastfeeding at least 1 month, and hormone therapy); effect-modifier variables (physical activity, BMI, and w/h ratio), when not evaluated as effect modifier variables, were adjusted as a covariate; when stratified via waist circumference, w/h ratio was not adjusted
  4. bParticipants stratified by BMI as non-obese (BMI < 30, n = 3675) or obese (BMI ≥ 30, n = 1704); interaction tests presented for the effect of BMI on the association between CRC and rs2191349 (effect size −0.10, p-value 0.70), rs4607517 (effect size −0.30, p-value 0.65), rs174550 (effect size −0.70, p-value 0.09), and rs11605924 (effect size −0.38, p-value 0.15)
  5. cParticipants stratified by waist circumference as non-obese (waist ≤88 cm; n = 3042) or obese (waist >88 cm; n = 2337); interaction tests presented for the effect of waist circumference on the association between CRC and rs2191349 (effect size −0.21, p-value 0.38), rs174550 (effect size −0.89, p-value 0.01), and rs11605924 (effect size −0.42, p-value 0.09)
  6. dParticipants stratified by w/h as non-obese (w/h ≤ 0.85; n = 3712) or obese (w/h > 0.85; n = 1667); interaction tests presented for the effect of w/h on the association between CRC and rs2191349 (effect size −0.21, p-value 0.40), rs10885122 (effect size 0.59, p-value 0.02), and rs174550 (effect size −0.34, p-value 0.34)
  7. eParticipants stratified by physical activity level as non-obese (MET ≥10; n = 2344) or obese (MET <10; n = 3035); interaction test presented for the effect of physical activity on the association between CRC and rs4607517 (effect size 0.94, p-value 0.14)
  8. fParticipants stratified by dietary fat intake as non-obese (< 40% calories from fat; n = 4325) or obese (≥ 40% calories from fat; n = 1054); interaction tests presented for the effect of dietary fat intake on the association between CRC and rs4607517 (effect size 1.45, p-value 0.01) and rs11558471 (effect size −1.53, p-value 0.04)